Key Takeaways
- A new Harvard study pinpoints human-caused climate change as a key driver of western U.S. wildfire activity over the last three decades, accounting for 65% of total fire emissions from 1997-2020.
- Nearly half of the exposure to harmful fine-particulate matter wildfire smoke, or PM2.5, from 1997-2020 is directly linked to climate change.
- The study highlights the need for stronger land-management strategies such as prescribed burning to mitigate extreme wildfires.
Across the western U.S., wildfires and the dangerous smoke that results have increased in frequency and intensity since the 1990s – that much is clear. Surprisingly less clear are the exact reasons why: While greenhouse gas-related global warming is often cited as a culprit, to what extent can this claim be quantified?
Atmospheric chemists and wildfire experts in the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS) have authored one of the most rigorous estimates to date of just how much of the fire damage and hazardous smoke of the last three decades is directly attributable to the warming temperatures and drier conditions caused by climate change.
Researchers led by Loretta Mickley, senior research fellow in chemistry-climate interactions and leader of the Atmospheric Chemistry Modeling Group at Harvard SEAS, report in Proceedings of the National Academy of Sciences that climate change directly accounts for 60-82% of total burned area in western U.S. forests and 33% in central and southern California since the early 1990s. On average, that’s 65% of total fire emissions in the U.S. between 1997 and 2020.
In turn, from 1997 to 2020, nearly half of the most dangerous types of wildfire smoke in the western U.S., what’s called fine particulate matter, or PM2.5, for its very small particle size and ability to penetrate lungs and the bloodstream, can be traced directly back to climate change. From 2010 to 2020, climate change explains 58% of the increase in this type of smoke pollution.
A combination of observations, machine learning models, and large climate models netted these conclusions, with the Harvard team’s analysis clearly showing how changing weather conditions and rising temperatures have affected the behavior of wildfires since the 1990s. The researchers also used a chemical transport model, GEOS-Chem, to estimate how much smoke PM2.5 is attributable to human-driven climate change.
“Our goal was to quantify how much climate change has amplified and exacerbated smoke exposure in the western U.S.,” Mickley said. “Our hope is that this work will spur efforts to think more deeply about how we manage land and wildfires in the western U.S., as well as greater interest toward doing something about the increase in greenhouse gases.”
To make their claims, the researchers first mapped a series of ecosystems across the western U.S., such as northwest forested mountains, mediterranean California, and the cold deserts of the interior, and compiled decades of records on weather, levels of vegetation, and areas burned in each. Machine learning models helped them deduce how temperature, humidity, and aridity of vegetation translated into fire activity.
The analysis confirmed that pollution from other sources, such as factories, dropped significantly -- around 44% -- from 1997 to 2020, highlighting the success of the Clean Air Act. Yet wildfire smoke has seen an opposing trend, increasing steadily during that time.
Hardest hit with wildfire smoke were areas that included northern California and parts of Oregon, Washington, and Idaho, where climate-driven smoke made up 44-66% of total PM2.5 between 2010 and 2020.
“If you lived in these areas from 2010 to 2020, at least half of all of the fine PM you were breathing came from smoke,” Mickley said.
In ongoing work, the team is trying to quantify how much the 20th-century legacy of fire suppression may have amplified the effects of climate change on wildfire activity today. Accumulated underbrush and increasingly dense forests have likely provided more fuel for these fires and therefore contributed to increased smoke exposures.
Mickley and others’ work continues to underscore the urgent need for land managers and communities in the western U.S. to deploy counteractive measures, notably prescribed burning in the most fire-prone areas. Prescribed burning can clear out underbrush and limit the spread of potentially catastrophic fires in the future.
The research was supported by the Modeling, Analysis, Prediction, and Projection program of the Climate Program Office, part of the National Oceanic and Atmospheric Administration (Grant No. NA22OAR4310140). The paper’s first author is former Harvard postdoctoral researcher Xu Feng. Co-authors are Jed O. Kaplan, Makoto Kelp, Yang Li, and Tianjia Liu.
Mickley will present the research at the Jacob Bjerknes Lecture, part of the American Geophysical Union’s December meeting.